From reduction to inflection

« previous post | next post »

Over the past dozen years, there's been a scattering of LLOG posts about various forms of a periphrastic future construction in English:

"I'ma", 7/3/2005
"I'monna", 7/3/2005
"'On' time", 8/4/20015
"Finna and tryna", 8/5/2005
"I'm a?", 9/19/2009
"I'ma stay with the youngsters", 5/14/2010
"Gonna, gone, onna, a — on?", 8/10/2012

But there's a quasi-inflectional aspect to this development that I don't think I've noticed before — and I haven't seen it discussed in the literature on this topic either, though I may well have missed the coverage. Specifically, some of the forms are apparently only available in the first person.

Like most Americans, I use the full historical form "going to" in formal writing but almost never in speech. The reduced form usually spelled "gonna" is a common option, for me as for others. A slightly less reduced form [gɔɪnə] (rhymes with "coin a") is possible. But in the first person, my normal usage is a more completely reduced form ['ɐɪmənə], in which the initial /g/ of "going to" is completely elided, and both of its vowels are completely unstressed and reduced.

This is sometimes spelled "I'monna", as in this passage from Carson Kreitzer's Self Defense and Other Plays:

FANNY LOU : You know I don't love you because you act like this sometimes.
ADAM: uh-huh.
FANNY LOU: I love you in spite of it.
ADAM: (smiling) Well, same goes for me. Double.

I'monna walk you home from every show.
ADAM: It'll be like we're courting again.
FANNY LOU: Did we stop?

The thing is, the form with elided /g/ isn't possible for me except in the first person — using the approximate spelling above:

I'monna try.
*You'reonna try.
*She'sonna try.
*We'reonna try.
*They'ronna try.

I recognize that it's always dangerous to study variation by intuitive judgment, but I'm confident that those asterisks represent at least a strong difference in frequency of usage. And FWIW, I really do feel a grammatical difference there.

This might be because the /m/ of "I'm" is more likely to elide a following /g/ by phonetic assimilation and reduction than the /r/ and /s/ of the other forms are. And no doubt that difference is the causal explanation for the development — though note that a purely phonetic form of this process might be more likely to produce ['ɐɪŋənə] than ['ɐɪmənə]. And the form with a velar nasal is certainly possible for me, though my admittedly unreliable sociolinguistic intuition says that it's less likely.

But just as the form spelled "gonna", though originally a phonetic reduction, is now clearly lexicalized (morpho-lexicalized?), the same thing seems to have happened with the form spelled "I'monna".

And I wonder whether there's something similar going on with the form commonly spelled "on". Thus in the first few pages of Kathryn Stockett's The Help we get:

Law, I'm on have to do it while the ladies is here, I guess.

Got to be the worst place in the world, inside a oven. You in here, you either cleaning or you getting cooked. Tonight I just know I'm on have that dream I'm stuck inside and the gas gets turned on.

And my work shoes is so thin, they look like they starving to death. New pair cost seven dollars though, which means I'm on be eating cabbage and tomato till I turn into Br'er Rabbit.

I been worried all afternoon, thinking bout Minny. I got to put it out a my mind if I'm on get some sleep tonight.

I know what happen between Constantine and Miss Skeeter's mama and ain't no way I'm on tell her that story.

I got a load a Mister Leefolt's shirts to iron and after this I'm on get a pot roast going.

But nowhere in that book could I find any instances of "he's on", "she's on", "we're on", "they're on" in this periphrastic future sense.

In an earlier post, I quoted from novelistic dialogue in which the "I'm on <VERB>" construction is attributed to various southerners. Looking back over the cited works, again I don't find any instances of "you're on", "(s)he's on", "it's on", "we're on", "they're on" as periphrastic futures.

Whether or not I'm right that the "I'monna" (and "I'm on") forms have been morphologized, this is exactly the sort of process that creates inflections over historical time.



  1. Yuval said,

    January 10, 2018 @ 8:54 pm

    The third-person equivalent might be "he gon ". In this case, it's the copula that gets elided, but maybe the process is the same?

    [(myl) Copula deletion is a different (and more general) process, as I understand it.]

  2. Joe Fineman said,

    January 10, 2018 @ 9:25 pm

    How is this related to the form encountered in black talk and usually spelled "ahmo"? It has been around for many years & AFAIK also lacks persons other than the first.

  3. Andrew Usher said,

    January 10, 2018 @ 11:22 pm

    All the same, surely – the form you spell "I'monna" is that used in standard speech, while "I'm on", "ahmo", and whatever else are representations of non-standard dialect. But I can fairly say we have a thing here that is only used in the first person singular. That's not much of an occasion for surprise given the way weak forms develope in languages – why this particular one, I could only suggest that "gonna" occurs more often in the 1sg. than anywhere else, and is thus most prone to weakening there. Further as cited here corresponding forms in the other persons sound awkward.

    By the way, I have always wanted to spell it "I'm'na", with two apostrophes. There is a bias against ever using two in a contraction in writing, but letting one stand for an extra syllable is no stranger than the same in "didn't" and such, which are normally two syllables though spelled with just one vowel.

    k_over_hbarc at

  4. John Lawler said,

    January 10, 2018 @ 11:48 pm

    In my speech the high front diphthongal gesture of I in I'm gonna is almost always reduced: /'a:mənə/. And, yes, it's restricted to first person. Nothing strange there, really, in a form that's undergoing change.

  5. Roger said,

    January 11, 2018 @ 1:11 am

    I hear "you'rena" occasionally. It's not as common, but it's definitely there in speech where I live in the DC area.

    Also "I'm'a". So, a sentence might look like, "You'rena do this, while I'm'a do that."

  6. phspaelti said,

    January 11, 2018 @ 4:11 am

    What about interaction with negation? Are there no negative forms?

  7. cliff arroyo said,

    January 11, 2018 @ 5:26 am

    @john lawler " /'a:mənə/ yes, this is my normal pronunciation too.

    Also I'm to me is /'a:mə/. Similarly, for me I'll is /'a:l/ I think the l might be different from the l in all

    @Roger yeah, 'you'rena' or 'hes'ana' etc don't sound impossible or wrong to me. I'm not sure if I use them (or if I've heard them much) but they certainly seem possible, if not common

    @phspaelti for me the negative blocks the elision of g

  8. Gene Buckley said,

    January 11, 2018 @ 9:49 am

    I can attest that my usual colloquial pronunciation of this sequence has the velar nasal that Mark cites, though with a monophthongal stressed vowel, as ['aŋənə], alongside ['aŋgənə]. I've never noticed myself saying a form with [m] such as ['amənə], and that feels a bit weird in my mouth despite its apparent frequency. (Of course I also say the more careful ['aɪmɡənə].) I presumably would have learned this pattern either in North Jersey (till age 7) or Southern Oregon (after that).

  9. SlideSF said,

    January 11, 2018 @ 12:49 pm

    I’m no linguist, so I don’t know the correct terminology, but it seems that it is more difficult to switch from the labial M sound to the velar G in I’m gonna ( or going to) than it is in, say, the R to G in You’re gonna. In other words, it has more to do with the mechanics of the mouth than it does the way it sounds. Without being judgmental (because I do it too), I’monna is just a more lazy or relaxed pronunciation. If I am going to go to the trouble of adding that G, I will probably go to the trouble of saying I’m going to (or at least I’m going da).

  10. Vulcan With a Mullet said,

    January 11, 2018 @ 4:08 pm

    Seems to me the negation for "I'ma" would be "I ain't gonna", which doesn't lend itself to any further easy contraction, but can at least be reduced to three syllables with nasalization of the last two…

  11. David L said,

    January 12, 2018 @ 3:36 pm

    For George H.W. Bush, as interpreted by Dana Carvey, the negative version was something like 'nagonna'

  12. Rodger C said,

    January 13, 2018 @ 12:10 pm


  13. cliff arroyo said,

    January 14, 2018 @ 4:52 am

    Trying to reduce 'I ain't gonna' as much as seems natural I end up with

    ãĩŋə (maybe with the ash vowel instead of ai)

    I'm fairly sure I don't actually ever say this but it seems natural and plausible.

  14. LTX said,

    January 17, 2018 @ 1:21 pm

    Don't underestimate the work done to popularize "I'ma" by the R. Kelly single from years back, "I'm a/I'ma Flirt (remix)".

RSS feed for comments on this post